Construction of Clinical Outcome Prediction Model for Chronic Obstructive Pulmonary Disease Patients with Pulmonary Embolism

XU Qiang; LI Zhijun; DUAN Na; QIN Yang

doi:10.3969/j.issn.1004-437X.2026.12.018

Abstract：Objective To analyze the influencing factors of clinical outcomes in patients with chronic obstructive pulmonary disease(COPD)complicated with pulmonary embolism(PE),and to construct a risk prediction model.Methods The clinical data of 132 patients with COPD combined with PE(modeling group)admitted to General Hospital of Hebi Coal Industry(Group)Co.,LTD from August 2019 to August 2022 were retrospectively analyzed.They were followed up for 2 years after treatment and divided into death group and survival group according to clinical outcome.The clinical data of the two groups were compared.Multivariate logistic regression was used to analyze the risk factors affecting the clinical outcome of COPD patients with PE,and the prediction model of clinical outcome of COPD patients with PE was constructed and verified internally.In addition,44 patients with COPD combined with PE(validation group)were selected as external validation of the prediction model according to a ratio of 3∶1.Results There was no significant difference in clinical data between the modeling group and the verification group(P＞0.05).Univariate analysis showed that age,pulmonary function grade,D-dimer,fibrinogen(FIB)level,platelet count and embolization degree were significantly associated with the clinical outcome of COPD patients with PE(P＜0.05).Multivariate logistic regression analysis showed that age,pulmonary function grade,D-dimer,FIB level,platelet count and embolization degree were independent risk factors affecting the clinical outcome of COPD patients with PE(P＜0.05).Internal and external verification showed that the average absolute error between the actual measured value of the calibration curve and the predicted value of the model was 0.030 and 0.025,respectively.The receiver operating characteristic(ROC)curve showed that the area under the curve(AUC)was 0.870 and 0.844,respectively.The decision analysis curve showed that when the predicted risk threshold was 0.01-0.90,the nomogram model showed significant clinical net benefit.Conclusion Age,pulmonary function grade,D-dimer,FIB level,platelet count and embolization degree are closely related to the clinical outcome of COPD patients with PE.The prediction model of clinical outcome of COPD patients with PE based on this model has good accuracy and differentiation,and has high clinical application value.